Spatial evapotranspiration, rainfall and land use data in water accounting. Part 2: Reliability of water accounting results for policy decisions in the Awash basin

Water Accounting Plus (WAC) is a framework that summarizes complex hydrological processes and water management issues in river basins. The framework is designed to use satellite-based measurements of land and water variables and processes as input data. A general concern associated with the use of satellite measurements is their accuracy. This study focuses on the impact of the error in remote sensing measurements on water accounting and information provided to policy makers. The Awash Basin in the central Rift Valley in Ethiopia is used as a case study to explore the reliability of WAC outputs, in the light of input data errors. The Monte Carlo technique was used for stochastic simulation of WAC outputs over a period of 3 yr. The results show that the stochastic mean of the majority of WAC parameters and performance indicators are within 5% deviation from the original WAC values based on one single calculation. Stochastic computation is proposed as a standard procedure for WAC water accounting because it provides the uncertainty bandwidth for every WAC output, which is essential information for sound decision-making processes. The majority of WAC parameters and performance indicators have a coefficient of variation (CV) of less than 20 %, which implies that they are reliable and provide consistent information on the functioning of the basin. The results of the Awash Basin also indicate that the utilized flow and basin closure fraction (the degree to which available water in a basin is utilized) have a high margin of error and thus a low reliability. As such, the usefulness of them in formulating important policy decisions for the Awash Basin is limited. Other river basins will usually have a more accurate assessment of the discharge in the river mouth.